Skip to content
Conference

Gearbox fault diagnosis and feature visualization based on multiscale convolution network with LSTM and attention mechanism

Sep 2026 · International Conference on Mechanical Engineering, Materials and Automation Technology · Vol 14355, pp. 1435524 - 1435524-7 · 0 citations · 8 references
Engineering

TL;DR

An end-to-end fault diagnosis model combining Multi-Scale Convolutional Neural Network, MSCNN, Long Short-Term Memory (LSTM) network and attention mechanism is proposed to improve diagnostic accuracy and model interpretability for gearbox fault diagnosis.

Abstract

Aiming at the problems that traditional gearbox fault diagnosis methods rely on manual feature extraction, have poor adaptability to non-stationary vibration signals, and are difficult to effectively capture multi-scale fault features and key time periods, this paper proposes an end-to-end fault diagnosis model combining Multi-Scale Convolutional Neural Network (MSCNN), Long Short-Term Memory (LSTM) network and attention mechanism. The model uses dual-branch multi-scale convolution to extract local and global fault features, employs bidirectional LSTM to model temporal dependencies, and introduces an attention mechanism to adaptively assign weights to critical fault segments, so as to improve diagnostic accuracy and model interpretability. Experiments are carried out on the Southeast University gearbox fault dataset, with a traditional fully connected network based on handcrafted time-frequency features as the baseline. The results show that proposed model achieves 99.42% classification accuracy for five gear states which is 9.12% higher than the baseline. The t-SNE visualization and clustering metrics verify that the learned features have better intra-class aggregation and inter-class separation. This model effectively overcomes the limitations of traditional methods and provides a high-precision and interpretable scheme for gearbox fault diagnosis.

View source

Similar papers

Conference Open access Sep 2026

Planetary gearbox fault diagnosis method based on improved AVOA optimized CNN-BiLSTM-Attention

An intelligent fault diagnosis method based on an improved African Vultures Optimization Algorithm-optimized CNN-BiLSTM-Attention model is proposed for planetary gearboxes with non-stationary vibration signals and complex fault features, demonstrating good diagnostic accuracy, stability, and generalization ability.

Jian-Han Chen, Hao-Yong Zhang, Wen-Bo Kang · 0 citations
Open access Sep 2026

Rolling bearing fault diagnosis method based on dual attention multi-scale convolutional network

A bearing fault diagnosis method based on the Dual Attention Multi-Scale Convolutional Neural Network that achieves a peak diagnostic accuracy of 99.69% in cross-condition tasks and maintains accuracy above 92% under −5 dB SNR, outperforming BiTCN-Transformer, MTFC-T, and CTCA in accuracy, recall, and F1-score, while a...

Ya-Mei Zhang, Xiao-Hu Liu · 0 citations
Open access Oct 2026

A novel intelligent fault diagnosis model for planetary gearboxes using wavelet-based synchroextracting transform-based deep learning networks

Abstract Planetary gearboxes are critical components in rotating machinery where timely fault diagnosis is essential for operational safety and reliability. However, diagnosing faults from vibration signals generated by these complex systems remains challenging due to their non-stationary nature and inherent complexity...

Dan Yao, Min Sun, Ya-Ya Wang et al. · 0 citations
Oct 2026

A novel bearing fault diagnosis approach combining F-VMD with CNN-BiLSTM and cross-attention

To enhance the reliability and accuracy of bearing fault diagnosis under challenging operating conditions, this study proposes a method that integrates a combined fast Fourier transform (FFT) and variational mode decomposition (VMD) preprocessing framework with a convolutional neural network (CNN)-bidirectional long sh...

Jie Li, Yu-Ying Wang, Jie Yang et al. · 0 citations
Open access Sep 2026

Multi-Scale 1DCNN Bearing Fault Diagnosis Method Based on First-Layer Wide Convolution Under Various Operating Conditions and Noise Backgrounds

Results show that this method demonstrates superior noise immunity and the ability to diagnose faults without differentiating between operating conditions, and improving the model's ability to represent features effectively.

Sen Mu, Xiong Hu, Bing Wang · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.